[Improvement] Avoid creating two fileset on the same storage location

### What would you like to be improved?

When creating the Fileset, we should avoid creating multiple Fileset on the same storage location.
For example:
```
case1:
fileset1 -> /tmp/location1
fileset2 -> /tmp/location1

case2:
fileset1 -> /tmp/location1
fileset2 -> /tmp/location1/fileset1
```

### How should we improve?

We may need a reverse index from the storage location to the Fileset Id.

Start by locating the Fileset creation entry point and the existing storage-location metadata handling. Trace how Fileset identifiers and locations are persisted, then define tests covering identical and nested locations; done means conflicting Filesets are rejected rather than created.

Written by the indexing model from the issue text.